than the lack of funding, than the publishing companies, than old-fashioned scientists, than PZ Meyers, more than Linux, more than Apple, more than anything I know.
Ok, let me explain then. But before that, a disclaimer: I own Windows 95, 98, XP and Vista, and this represents only my opinion, not the opinion of my employers, co-workers and family.
And why is MSFT more detrimental to Science than anything else? It’s simple: the openness of their system. Maybe open is not the right word, transparency might better to explain it. As MSFT has the leadership on the OS market, vendors tend to attach their products (in this case laboratory equipments, analysis software, etc) to the OS. And how that is bad? Because when it works, it works fine, but when it doesn’t work, it’s a nightmare. Software registry that fails, legacy software that doesn’t work even when compatibility mode, no control over installation protocols. Most of the time you also don’t have free/open source tools that help you solver the problem.
After all this, you have to throw in license costs, expensive certification for sys admin in order to manage the machines, time wasted troubleshooting applications, OS, updates, etc. All of this drag all the way up and all the way down the Scientific ladder, increasing costs all over the board, and man-months of work that could be used in different aspects of research.
I’m not advocating either OS X, Linux, BSD or Solaris, I’m just saying that these systems are easier to use, cheaper (maybe not OS X) and provide a better experience to the users. And I’m not advocating that “regular” users should migrate to these systems, I’m just saying that if you are cost conscious scientist, they are a better alternative. Maybe if every university had an introduction to Linux/Unix course, maybe we would be able to cut costs every where, have better scientists and work more efficiently with our computers.
Do you think universities would have a course like this? I don’t think so, but it would be nice to see.